From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 5.5; Windows NT 4.0) Description of problem: I am trying to use RedHat Linux 7.1 on a Dell 8450 4 x 550MHz CPU's with 1GB of RAM. The system was previously running RH 6.2. After upgrading to 7.1, I lost the ability to connect to the network. The network card is the standard dual port card shipped with Dell servers. The network card looks like an OEM version of the Intel dual port 10/100 card. You can see the board assembly information in the "dmesg" output below. After trying many things, we found that we can get the card to work if we do these things: (1) connect both ports of the card to a network (2) either cold boot the system or boot after pressing the reset switch If both of the steps above are not performed, you cannot get access to the network. In this condition, the lights for one or both ports may be out. "ifconfig" reports the ports as UP, but the connection does not work. This is not an acceptable workaround because we need to be able to remotely re-boot the computer, and cannot do this if we must cycle power, or use the reset switch. We tried using a newer version of the same network card from Dell and observed the same behavior. I have included output from "dmesg" below. This is from a boot where the above 2 steps were followed, and our 1 desired network port is working. eepro100.c:v1.09j-t 9/29/99 Donald Becker http://cesdis.gsfc.nasa.gov/linux/drivers/eepro100.html eepro100.c: $Revision: 1.36 $ 2000/11/17 Modified by Andrey V. Savochkin <saw.com.sg> and others eth0: OEM i82557/i82558 10/100 Ethernet, 00:90:27:D5:CA:17, I/O at 0x2000, IRQ 54. Receiver lock-up bug exists -- enabling work-around. Board assembly 751110-001, Physical connectors present: RJ45 Primary interface chip i82555 PHY #1. General self-test: passed. Serial sub-system self-test: passed. Internal registers self-test: passed. ROM checksum self-test: passed (0x24c9f043). Receiver lock-up workaround activated. eth1: OEM i82557/i82558 10/100 Ethernet, 00:90:27:D5:CA:18, I/O at 0x2020, IRQ 53. Receiver lock-up bug exists -- enabling work-around. Board assembly 751110-001, Physical connectors present: RJ45 Primary interface chip i82555 PHY #1. General self-test: passed. Serial sub-system self-test: passed. Internal registers self-test: passed. ROM checksum self-test: passed (0x24c9f043). Receiver lock-up workaround activated. Again, this card was working properly under RH 6.2 when only 1 port is connected to the network (our desired method of operation). The card also works properly under NT 4 and Win2K. Any ideas about how to fix this, or who to contact, that may be able to fix this, would be greatly appreciated. How reproducible: Always Steps to Reproduce: 1. Connect only ethernet cable to the dual-port card 2. Boot the system 3. Network is not accessible Actual Results: The network is not accessible. ifconfig report the network as UP. ping cannot get out of the host. Other hosts cannot access this host. Some, or all of the LED's on the network card may be out. Expected Results: Access to the network should work. Additional info: Included in the description.
More of a driver issue, changing to kernel.
Could you try the e100 driver instead ? (eg edit /etc/modules.conf and replace eepro100 with e100 there)
Closed: Inactive